<![CDATA[Self-medication for cough is a common practice among non-health students. Inappropriate self-medication behavior may increase the risk of medication errors, highlighting the need for a statistical approach capable of modeling behavioral categories simultaneously. This study aimed to analyze cough self-medication behavior based on knowledge level using multinomial logistic regression among non-health students. An analytical observational study with a cross-sectional design was conducted involving 233 non-health students. The dependent variable was cough self-medication behavior categorized into several levels, while the independent variable was the level of knowledge. Statistical analysis was performed using multinomial logistic regression with maximum likelihood estimation, with the lowest behavior category set as the reference. Model significance was assessed using the likelihood ratio test, and model performance was evaluated through pseudo R-square values. The results showed that knowledge level was not significantly associated with cough self-medication behavior. The likelihood ratio test yielded a p-value of 0.100, indicating that the overall model was not statistically significant. The pseudo R-square value of 0.012 suggested that knowledge level explained only 1.2% of the variance in cough self-medication behavior. The estimated odds ratios indicated a tendency toward a lower likelihood of inappropriate self-medication behavior among students with higher knowledge levels; however, these associations were not statistically significant (p > 0.05). In conclusion, although knowledge level demonstrates a theoretically plausible direction of association, it is not a dominant determinant of cough self-medication behavior among non-health students. Further studies are recommended to include additional variables such as attitudes, prior medication experience, and access to health information to improve the explanatory power of the model.]]>
Copyrights © 2025
